Friday November 4 2011
Results of studies in mice may not apply to humans
“Eating overcooked meat is twice as likely to cause cancer than previously thought,” warned the front page of the Daily Express today.
The headline was based on the results of an animal study, in which mice were genetically modified to produce human versions of enzymes called sulphotransferases. These enzymes break down various drugs and other substances. The researchers found that human sulphotransferases in mice that were genetically predisposed to developing tumours led to an increase in the number and frequency of colon tumours after the mice were treated with a substance called PhIP. PhIP is formed when meat and fish are fried or grilled at high temperatures.
The results of this study were interpreted by the media to mean that overcooked or burnt meat could increase your risk of cancer. However, as the researchers point out, there are many differences between mice and men. Therefore, it is unclear how relevant the findings of this study are for human health. PhIP is listed as a class 2B carcinogen (“possibly carcinogenic to humans”) by the International Agency for Research on Cancer. However, further research will need to establish whether PhIP does cause cancer in humans.
Where did the story come from?
The study was carried out by researchers from the Norwegian Institute of Public Health and the German Institute of Human Nutrition. It was funded by the Norwegian Research Council.
The study was published in the peer-reviewed journal Molecular Carcinogenesis.
The Daily Express and Daily Mail reported this story. While the results of the study and the conclusions of the researchers were reported accurately in both news reports, they also put too much emphasis on human cancer risk. The Express’s article also included useful information from Cancer Research UK about how this research question could be best addressed in humans.
What kind of research was this?
This animal study aimed to determine whether the production of certain enzymes present in humans would change the carcinogenic effect of two substances. Humans and mice have different enzymes in different parts of the body. In this study, the researchers created genetically modified mice that produced human versions of enzymes called sulphotransferases. This group of enzymes breaks down certain drugs and other substances in the body.
Mice are often used to test whether compounds are harmful to humans. This is because such experiments can be performed quickly and because it would be unethical to carry out trials in humans using potentially harmful substances. However, although such experiments in mice are useful, they have limitations in that the results may not apply to human health.
What did the research involve?
The researchers bred four types of mice:
- wild-type mice (WT or “normal” mice)
- mice that were genetically modified to produce human sulphotransferases (hSULT mice)
- mice that were genetically predisposed to developing tumours (Min mice)
- mice that were genetically predisposed to developing tumours and that produced human sulphotransferases (Min/hSULT mice)
They then tested the effect of giving the mice two compounds. HMF is a compound that is generated at moderate temperatures in foods containing sugars. PhIP is a compound that is formed when meat and fish are fried or grilled at high temperatures.
The mice were fed a low dose of HMF (375mg/kg of body weight), a high dose of HMF (750mg/kg body weight) or salt water three times a week for 11 weeks to test the effect of HMF. Other mice received injections of either 50mg/kg body weight of PhIP or salt water one week before they were born and one, two and three weeks after birth.
The presence of tumours and tumour size were then recorded. The researchers compared the number and incidence of tumours in the different mice that were fed the different compounds.
What were the basic results?
HMF did not affect the formation of tumours.
Treatment with PhIP increased the formation of tumours in Min and Min/hSULT mice, which were predisposed to developing tumours. However, PhIP had no significant effect on tumour development in WT or hSULT mice.
Min/hSULT mice treated with PhIP had three times as many tumours in the colon and a higher incidence of colon cancer compared to Min mice that were treated with PhIP. Min mice treated with PhIP had 0.4 colon tumours on average, compared to 1.3 tumours in Min/hSULT mice. The incidence of colon cancer was 31% in Min mice, compared to 80% in Min/hSULT mice. However, there was no difference in the number or incidence of tumours in the small intestine, or of “aberrant crypt foci” (clusters of abnormal tube-like glands which might lead to cancer) in the colon.
How did the researchers interpret the results?
The researchers concluded that their results show that “Min/hSULT mice are more sensitive to tumour development in the colon after PhIP treatment than conventional Min mice.” The researchers also said that “humans may be more sensitive than mice” towards certain compounds, and “this should be accounted for in risk assessments based on rodent data”.
In this study, mice were genetically modified to produce human versions of enzymes called sulphotransferases. The researchers found that the production of human sulphotransferases in mice that were predisposed to developing tumours increased the number and incidence of colon tumours after they were treated with a substance called PhIP. PhIP is formed when meat and fish are fried or grilled at high temperatures. The International Agency for Research on Cancer lists PhIP as a class 2B carcinogen (“possibly carcinogenic to humans”).
The newspapers interpreted the results to mean that overcooked or burnt meat could increase your risk of cancer. However, as the researchers point out, there are many differences between mice and humans. It is not clear how relevant the findings are for human health, especially as PhIP did not lead to tumour development in healthy mice that produced human sulphotransferases but were not genetically susceptible to tumours.
Large cohort studies, which follow people up for a long period, would give the best evidence for the effects of PhIP on humans. Exposing people to charred food compounds in a randomised controlled trial would be difficult to do for long periods, and would be potentially unethical as the substances produced are possible carcinogens. At least two published cohort studies have shown that methods of cooking meat do not affect the risk of lung or prostate cancer.